*****Table 1: Model 1 ANEP
nbreg anep lag_anep nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)

*****Table 1: Model 2 ENEP
reg enep lag_enep nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)

*****Figure 3: Simulation using CLARIFY
estsimp reg enep lag_enep nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)
setx mean
setx lag_enep 10
setx nspr 4
simqi

estsimp reg enep lag_enep nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)
setx mean
setx lag_enep 5
setx nspr 4
simqi

*****Endnote#8: Analyses using legislative seats data
nbreg anlp lag_anlp nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)
reg enlp lag_enlp nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)

*****Endnote#9: Analyses using a new ENEP
reg enep1pc lag_enep1pc nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)

*****Endnote#11: Analyses using Carey and Hix (2011) data
reg carey_enep lag_carey_enep nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)
reg carey_enlp lag_carey_enlp nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, robust cluster(countryid)

****ROBUSTNESS CHECKS****

*****Detecting outliers
corr enep nspr, means
reg enep nspr
predict resid, residual
predict stdresid, rstandard
predict rstudent, rstudent
predict cooksd, cooksd
extremes stdresid rstudent cooksd
***(outliers if stdresid is larger than 3, and cooksd is larger than 4/N, which is 4/123=0.0325)

***Re-analysis with inclusion of four outliers
reg enep lag_enep nspr logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies ecua84 ecua86 brz06 brz10, robust cluster(countryid)

****Instrumental variable approach
ivregress 2sls enep lag_enep (nspr=regis_den) logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, first robust cluster(countryid)
ivregress 2sls enlp lag_enlp (nspr=regis_den) logadm fearon fearonlogadm spatial runoff concurrent preferential lgdpgrowth loglinflation foundingyr dem_years subsidies, first robust cluster(countryid)

